专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20120221330A1 LEVERAGING SPEECH RECOGNIZER FEEDBACK FOR VOICE ACTIVITY DETECTION 有权
标题翻译：引导语音识别器反馈进行语音活动检测
公开(公告)号：US20120221330A1
公开(公告)日：2012-08-30
申请号：US13035044
申请日：2011-02-25
申请人： Albert Joseph Kishan Thambiratnam , Weiwu Zhu , Frank Torsten Bernd Seide
发明人： Albert Joseph Kishan Thambiratnam , Weiwu Zhu , Frank Torsten Bernd Seide
IPC分类号： G10L15/26 , G10L15/00
CPC分类号： G10L25/84 , G10L15/04 , G10L15/26 , G10L25/45
摘要： A voice activity detection (VAD) module analyzes a media file, such as an audio file or a video file, to determine whether one or more frames of the media file include speech. A speech recognizer generates feedback relating to an accuracy of the VAD determination. The VAD module leverages the feedback to improve subsequent VAD determinations. The VAD module also utilizes a look-ahead window associated with the media file to adjust estimated probabilities or VAD decisions for previously processed frames.
摘要翻译：语音活动检测（VAD）模块分析诸如音频文件或视频文件的媒体文件，以确定媒体文件的一个或多个帧是否包含语音。语音识别器产生与VAD确定的精度有关的反馈。 VAD模块利用反馈来改进随后的VAD确定。 VAD模块还利用与媒体文件相关联的预览窗口来调整对先前处理的帧的估计概率或VAD决定。

2. 发明授权

US08650029B2 Leveraging speech recognizer feedback for voice activity detection 有权
标题翻译：利用语音识别器反馈进行语音活动检测
公开(公告)号：US08650029B2
公开(公告)日：2014-02-11
申请号：US13035044
申请日：2011-02-25
申请人： Albert Joseph Kishan Thambiratnam , Weiwu Zhu , Frank Torsten Bernd Seide
发明人： Albert Joseph Kishan Thambiratnam , Weiwu Zhu , Frank Torsten Bernd Seide
IPC分类号： G10L15/00
CPC分类号： G10L25/84 , G10L15/04 , G10L15/26 , G10L25/45
摘要： A voice activity detection (VAD) module analyzes a media file, such as an audio file or a video file, to determine whether one or more frames of the media file include speech. A speech recognizer generates feedback relating to an accuracy of the VAD determination. The VAD module leverages the feedback to improve subsequent VAD determinations. The VAD module also utilizes a look-ahead window associated with the media file to adjust estimated probabilities or VAD decisions for previously processed frames.
摘要翻译：语音活动检测（VAD）模块分析诸如音频文件或视频文件的媒体文件，以确定媒体文件的一个或多个帧是否包含语音。语音识别器产生与VAD确定的精度有关的反馈。 VAD模块利用反馈来改进随后的VAD确定。 VAD模块还利用与媒体文件相关联的预览窗口来调整对先前处理的帧的估计概率或VAD决定。

3. 发明申请

US20120226696A1 Keyword Generation for Media Content 有权
标题翻译：媒体内容的关键字生成
公开(公告)号：US20120226696A1
公开(公告)日：2012-09-06
申请号：US13040640
申请日：2011-03-04
申请人： Albert Joseph Kishan Thambiratnam , Sha Meng , Gang Li , Frank Torsten Bernd Seide
发明人： Albert Joseph Kishan Thambiratnam , Sha Meng , Gang Li , Frank Torsten Bernd Seide
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： G06F17/30796 , G06F17/30864
摘要： In various embodiments, a transcript that represents a media file is created. Keyword candidates that may represent topics and/or content associated with the media content are then be extracted from the transcript. Furthermore, a keyword set may be generated for the media content utilizing a mutual information criteria. In other embodiments, one or more queries may be generated based at least in part on the transcript, and a plurality of web documents may be retrieved based at least in part on the one or more queries. Additional keyword candidates may be extracted from each web document and then ranked. A subset of the keyword candidates may then be selected to form a keyword set associated with the media content.
摘要翻译：在各种实施例中，创建表示媒体文件的抄本。然后可以从成绩单中提取可能表示与媒体内容相关联的主题和/或内容的关键字候选。此外，可以利用互信息标准为媒体内容生成关键字集合。在其他实施例中，可以至少部分地基于抄本生成一个或多个查询，并且可以至少部分地基于一个或多个查询来检索多个web文档。可以从每个web文档中提取附加的关键字候选，然后排序。然后可以选择关键字候选的子集以形成与媒体内容相关联的关键字集合。

4. 发明授权

US09483557B2 Keyword generation for media content 有权
标题翻译：媒体内容的关键字生成
公开(公告)号：US09483557B2
公开(公告)日：2016-11-01
申请号：US13040640
申请日：2011-03-04
申请人： Albert Joseph Kishan Thambiratnam , Sha Meng , Gang Li , Frank Torsten Bernd Seide
发明人： Albert Joseph Kishan Thambiratnam , Sha Meng , Gang Li , Frank Torsten Bernd Seide
IPC分类号： G06F17/30
CPC分类号： G06F17/30796 , G06F17/30864
摘要： In various embodiments, a transcript that represents a media file is created. Keyword candidates that may represent topics and/or content associated with the media content are then be extracted from the transcript. Furthermore, a keyword set may be generated for the media content utilizing a mutual information criteria. In other embodiments, one or more queries may be generated based at least in part on the transcript, and a plurality of web documents may be retrieved based at least in part on the one or more queries. Additional keyword candidates may be extracted from each web document and then ranked. A subset of the keyword candidates may then be selected to form a keyword set associated with the media content.
摘要翻译：在各种实施例中，创建表示媒体文件的抄本。然后可以从成绩单中提取可能表示与媒体内容相关联的主题和/或内容的关键字候选。此外，可以利用互信息标准为媒体内容生成关键字集合。在其他实施例中，可以至少部分地基于抄本生成一个或多个查询，并且可以至少部分地基于一个或多个查询来检索多个web文档。可以从每个web文档中提取附加的关键字候选，然后排序。然后可以选择关键字候选的子集以形成与媒体内容相关联的关键字集合。

5. 发明申请

US20130191126A1 Subword-Based Multi-Level Pronunciation Adaptation for Recognizing Accented Speech 有权
标题翻译：基于字的多层次发音适应识别口音
公开(公告)号：US20130191126A1
公开(公告)日：2013-07-25
申请号：US13355215
申请日：2012-01-20
申请人： Albert Joseph Kishan Thambiratnam , Timo Pascal Mertens , Frank Torsten Bernd Seide
发明人： Albert Joseph Kishan Thambiratnam , Timo Pascal Mertens , Frank Torsten Bernd Seide
IPC分类号： G10L15/06
CPC分类号： G10L15/187 , G10L15/06
摘要： Techniques are described for training a speech recognition model for accented speech. A subword parse table is employed that models mispronunciations at multiple subword levels, such as the syllable, position-specific cluster, and/or phone levels. Mispronunciation probability data is then generated at each level based on inputted training data, such as phone-level annotated transcripts of accented speech. Data from different levels of the subword parse table may then be combined to determine the accented speech model. Mispronunciation probability data at each subword level is based at least in part on context at that level. In some embodiments, phone-level annotated transcripts are generated using a semi-supervised method.
摘要翻译：描述了用于训练重音语音的语音识别模型的技术。使用子字解析表来对多个子词级别（例如音节，位置特定的群集和/或电话层级）进行误导。然后基于输入的训练数据（例如重音语音的电话级注释誊本）在每个级别生成分秒传播概率数据。然后可以组合来自子词解析表的不同级别的数据以确定重音语音模型。每个子词级别的通配概率数据至少部分地基于该级别的上下文。在一些实施例中，使用半监督方法生成电话级注释的记录。

6. 发明授权

US08825481B2 Subword-based multi-level pronunciation adaptation for recognizing accented speech 有权
标题翻译：用于识别重音语音的基于字的多级发音适应
公开(公告)号：US08825481B2
公开(公告)日：2014-09-02
申请号：US13355215
申请日：2012-01-20
申请人： Albert Joseph Kishan Thambiratnam , Timo Pascal Mertens , Frank Torsten Bernd Seide
发明人： Albert Joseph Kishan Thambiratnam , Timo Pascal Mertens , Frank Torsten Bernd Seide
IPC分类号： G10L15/00 , G10L15/06
CPC分类号： G10L15/187 , G10L15/06
摘要： Techniques are described for training a speech recognition model for accented speech. A subword parse table is employed that models mispronunciations at multiple subword levels, such as the syllable, position-specific cluster, and/or phone levels. Mispronunciation probability data is then generated at each level based on inputted training data, such as phone-level annotated transcripts of accented speech. Data from different levels of the subword parse table may then be combined to determine the accented speech model. Mispronunciation probability data at each subword level is based at least in part on context at that level. In some embodiments, phone-level annotated transcripts are generated using a semi-supervised method.
摘要翻译：描述了用于训练重音语音的语音识别模型的技术。使用子字解析表来对多个子词级别（例如音节，位置特定的群集和/或电话层级）进行误导。然后基于输入的训练数据（例如重音语音的电话级注释誊本），在每个级别生成分数概率数据。然后可以组合来自子词解析表的不同级别的数据以确定重音语音模型。每个子词级别的通配概率数据至少部分地基于该级别的上下文。在一些实施例中，使用半监督方法生成电话级注释的记录。

7. 发明授权

US08700552B2 Exploiting sparseness in training deep neural networks 有权
标题翻译：在深层神经网络训练中利用稀疏性
公开(公告)号：US08700552B2
公开(公告)日：2014-04-15
申请号：US13305741
申请日：2011-11-28
申请人： Dong Yu , Li Deng , Frank Torsten Bernd Seide , Gang Li
发明人： Dong Yu , Li Deng , Frank Torsten Bernd Seide , Gang Li
IPC分类号： G06F15/18 , G06N3/08
CPC分类号： G06N3/08
摘要： Deep Neural Network (DNN) training technique embodiments are presented that train a DNN while exploiting the sparseness of non-zero hidden layer interconnection weight values. Generally, a fully connected DNN is initially trained by sweeping through a full training set a number of times. Then, for the most part, only the interconnections whose weight magnitudes exceed a minimum weight threshold are considered in further training. This minimum weight threshold can be established as a value that results in only a prescribed maximum number of interconnections being considered when setting interconnection weight values via an error back-propagation procedure during the training. It is noted that the continued DNN training tends to converge much faster than the initial training.
摘要翻译：提出了深层神经网络（DNN）训练技术实施例，其训练DNN，同时利用非零隐藏层互连权重值的稀疏性。通常，完全连接的DNN最初通过遍历完整的训练集多次进行训练。那么，在大多数情况下，只有重量大小超过最小重量阈值的互连在进一步的训练中被考虑。该最小权重阈值可以被建立为在训练期间通过错误反向传播过程设置互连权重值时仅考虑规定的最大数量的互连的值。值得注意的是，继续进行的DNN训练往往比初始训练快得多。

8. 发明授权

US07617104B2 Method of speech recognition using hidden trajectory Hidden Markov Models 失效
标题翻译：使用隐藏轨迹隐马尔可夫模型的语音识别方法
公开(公告)号：US07617104B2
公开(公告)日：2009-11-10
申请号：US10348192
申请日：2003-01-21
申请人： Li Deng , Jian-Iai Zhou , Frank Torsten Bernd Seide
发明人： Li Deng , Jian-Iai Zhou , Frank Torsten Bernd Seide
IPC分类号： G10L15/14
CPC分类号： G10L15/142
摘要： A method of speech recognition is provided that determines a production-related value, vocal-tract resonance frequencies in particular, for a state at a particular frame based on the production-related values associated with two preceding frames using a recursion. The production-related value is used to determine a probability distribution of the observed feature vector for the state. A probability for an observed value received for the frame is then determined from the probability distribution. Under one embodiment, the production-related value is determined using a noise-free recursive definition for the value. Use of the recursion substantially improves the decoding speed. When the decoding algorithm is applied to training data with known phonetic transcripts, forced alignment is created which improves the phone segmentation obtained from the prior art.
摘要翻译：提供了一种语音识别方法，其基于与使用递归的两个先前帧相关联的生产相关值来确定特定帧处的状态的特定生产相关值，声道共振频率。生产相关值用于确定观察到的状态特征向量的概率分布。然后从概率分布确定为帧接收的观测值的概率。在一个实施例中，使用针对该值的无噪声递归定义来确定生产相关值。使用递归大大提高了解码速度。当解码算法应用于具有已知语音抄本的训练数据时，产生强制对准，其改善了从现有技术获得的电话分割。

9. 发明申请

US20130138436A1 DISCRIMINATIVE PRETRAINING OF DEEP NEURAL NETWORKS 有权
标题翻译：深层神经网络的辨别性预测
公开(公告)号：US20130138436A1
公开(公告)日：2013-05-30
申请号：US13304643
申请日：2011-11-26
申请人： Dong Yu , Li Deng , Frank Torsten Bernd Seide , Gang Li
发明人： Dong Yu , Li Deng , Frank Torsten Bernd Seide , Gang Li
IPC分类号： G10L15/16
CPC分类号： G06N3/08 , G06N3/04
摘要： Discriminative pretraining technique embodiments are presented that pretrain the hidden layers of a Deep Neural Network (DNN). In general, a one-hidden-layer neural network is trained first using labels discriminatively with error back-propagation (BP). Then, after discarding an output layer in the previous one-hidden-layer neural network, another randomly initialized hidden layer is added on top of the previously trained hidden layer along with a new output layer that represents the targets for classification or recognition. The resulting multiple-hidden-layer DNN is then discriminatively trained using the same strategy, and so on until the desired number of hidden layers is reached. This produces a pretrained DNN. The discriminative pretraining technique embodiments have the advantage of bringing the DNN layer weights close to a good local optimum, while still leaving them in a range with a high gradient so that they can be fine-tuned effectively.
摘要翻译：提出了预先训练深层神经网络（DNN）的隐藏层的识别性预训练技术实施例。一般来说，首先使用带有误差反向传播（BP）的标签对标签进行单层隐藏层神经网络的训练。然后，在丢弃前一个隐藏层神经网络中的输出层之后，将另一个随机初始化的隐藏层与先前训练过的隐藏层一起添加，并将其代表表示用于分类或识别的目标的新输出层。然后使用相同的策略对所得到的多隐层DNN进行鉴别性训练，等等，直到达到所需数量的隐藏层。这产生了一个预训练的DNN。鉴别预培训技术实施例具有使DNN层权重接近良好的局部最优值的优点，同时仍将它们留在具有高梯度的范围内，使得它们可以被有效地微调。

10. 发明授权

US07693267B2 Personalized user specific grammars 失效
标题翻译：个性化用户特定语法
公开(公告)号：US07693267B2
公开(公告)日：2010-04-06
申请号：US11324208
申请日：2005-12-30
申请人： David Andrew Howell , Sridhar Sundararaman , David T. Fong , Frank Torsten Bernd Seide
发明人： David Andrew Howell , Sridhar Sundararaman , David T. Fong , Frank Torsten Bernd Seide
IPC分类号： H04M11/00
CPC分类号： H04M3/5307 , H04M2201/405 , H04M2201/60 , H04M2203/253 , H04M2203/4509
摘要： Improved systems and methods are provided for transcribing audio files of voice mails sent over a unified messaging system. Customized grammars specific to a voice mail recipient are created and utilized to transcribe a received voice mail by comparing the audio file to commonly utilized words, names, acronyms, and phrases used by the recipient. Key elements are identified from the resulting text transcription to aid the recipient in processing received voice mails based on the significant content contained in the voice mail.
摘要翻译：提供了改进的系统和方法，用于转录通过统一消息系统发送的语音邮件的音频文件。通过将音频文件与收件人使用的常用词，名称，首字母缩略词和短语进行比较，创建和利用特定于语音邮件收件人的定制语法来转录接收到的语音邮件。从所得到的文本转录中识别关键要素，以帮助接收者基于语音邮件中包含的重要内容来处理接收的语音邮件。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式